The HlV-1 protease inhibitors (Pis) in highly active anti-retroviral therapy (HAART) can effectively suppress plasma viral load. However, HIV-1 persists in sanctuary sites due to sub-therapeutic concentrations of Pis. In the central nervous system (CNS) reservoir and in the blood-brain brain (BBB) endothelial cells (ECs) the Pglycoprotein (P-gp) efflux pumps are implicated in decreasing PI levels. Recent findings have documented P-gp expression in vascular ECs as well. In vitro studies using peripheral blood mononuclear cells (PBMCs) from HIV-1 infected patients have also shown their increased trans-endothelial migration (TEM) potential and have demonstrated their ability to propagate and transmit virus, suggesting their role in forming perivascular reservoirs of HIV-1. P-gp expression in vascular ECs and transmigratory PBMCs may allow HAART inefficacy due to decreased levels of Pis in sub-endothelial compartments. Our preliminary data show that human aortic endothelial cells (HAECs) express functional P-gp and exposure to the HIV-1 Tat protein or TNF-alpha further enhanced cell surface levels of P-gp. Our previous studies showed that CD83+ dendritic progenitors can harbor HIV-1 and can transmit the virus to replication competent T-cells. Recent reports have shown that certain CD83 PBMC population are infectable and possess enhanced TEM potential. The role of P-gp efflux pumps in modulating the anti-HIV activity of Pis in both vascular ECs and transmigratory PBMCs has not been investigated. We are therefore proposing to test a novel hypothesis that an enhanced expression of P-gp in the vascular ECs and in HIV-1 infected transmigratory PBMCs efflux the HIV-1 Pis and allows perivascular propagation and transmission of HIV-1 in the sub-endothelial compartment. By utilizing an in vitro model of the sub-endothelial compartment containing HIV-1 infected PBMCs. We aim to monitor the efficacy of Pis (amprenavir, lopinavir, nelfinavir or saquinavir) in suppressing HTV-1 propagation in transmigrated PBMCs. The modulatory effects of HTV-1 infection and/or HTV-1 induced factors (e.g. TNF-alpha and/or Tat) on P-gp expression and efflux function will also be monitored. A direct role of P-gp in decreasing intracellular PI levels hi both these cell types will be established by using pharmacologic inhibitors of Pgp and/or cells transfected with P-gp specific small interference RNAs (siRNAs). The following specific aims will be carried out to achieve our goals: (i) to monitor the effects of HIV-1 infection on P-gp expression and efflux function, (ii) to determine the role of P-gp hi facilitating perivascular reservoirs for HTV-1, and (iii) to investigate whether P-gp suppression will enhance Pi-efficacy in perivascular reservoirs for HTV-1. The proposed studies are expected to provide exploratory new information on existence of a perivascular endothelial barrier that may compromise the effectiveness of Pis and allow development of an HTV-1 sanctuary.